The long-term objective of the work proposed here is to develop and demonstrate Random Hand Hygiene Prompts (RHHP). RHHP is a device that 1) plays computer generated voice messages over loudspeakers at a hospital nursing station and 2) records hand hygiene events with soap and with sanitizer there. RHHP's voice messages play at random times and encourage healthcare workers to perform good hand hygiene. Phase I showed that listeners respond by washing their hands more often when RHHP's voice messages play than in their absence. In Phase II we will install RHHP in several hospital Intensive Care Units, where we will combine the hospital's census and RHHP's hand hygiene events to calculate hand hygiene events per patient per day, the standard measure of hand hygiene compliance. We will also measure the hospital acquired infection (HAI) rate. We will compare the hand hygiene compliance and the HAI rate in a baseline phase, when no voice messages play, to those rates in an intervention phase, when voice messages play. We will demonstrate that not only does hand hygiene increase in the intervention phase, but that the hospital acquired infection rate decreases as well. The result will be significant savings in lives and financial costs to the hospital. Further, RHHP reduces HAIs without using antibiotics, so it contributes to public health by minimizing the development of antibiotic-resistant organisms. We will optimize RHHP's hardware, software, phase lengths and message delivery parameters in Phase II, and investigate several research topics that can only be considered with the help of RHHP's detailed hand hygiene compliance measurements. At the same time, our collaborator Renoir Group will help us offer a simple version of RHHP for sale to customers convinced of its value by our Phase I results. In Phase III our collaborator A.C.C. Systems, Inc. will help us manufacture and market a more sophisticated RHHP to hospital chains. PUBLIC HEALTH RELEVANCE: Random Hand Hygiene Prompts will reduce the hospital acquired infection rate in hospitals where its voice messages play. As a result, fewer hospital patients will be colonized with infections on discharge and fewer infections will arise in the general public. Random Hand Hygiene Prompts will also reduce hospitals dependence on antibiotics. Then fewer strains of antibiotic-resistant bacteria will emerge, making it possible to treat more infections in the public with antibiotics